Skip to content

Commit

Permalink
Tomasz/tesseract52 (#19)
Browse files Browse the repository at this point in the history
* new package tesseract52

* build new package
  • Loading branch information
tomasz-h2o committed Jun 25, 2024
1 parent 30aea82 commit a8dd35c
Show file tree
Hide file tree
Showing 3 changed files with 156 additions and 0 deletions.
1 change: 1 addition & 0 deletions .github/workflows/build.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -18,3 +18,4 @@ jobs:
docker run --privileged --rm -v $(pwd)/packages:/work/packages -v $(pwd)/libspatialindex.yaml:/work/libspatialindex.yaml cgr.dev/chainguard/melange build --arch=x86_64 libspatialindex.yaml
docker run --privileged --rm -v $(pwd)/packages:/work/packages -v $(pwd)/mesa22.yaml:/work/mesa22.yaml cgr.dev/chainguard/melange build --arch=x86_64 mesa22.yaml
docker run --privileged --rm -v $(pwd)/packages:/work/packages -v $(pwd)/libjpeg-turbo2.yaml:/work/libjpeg-turbo2.yaml cgr.dev/chainguard/melange build --arch=x86_64 libjpeg-turbo2.yaml
docker run --privileged --rm -v $(pwd)/packages:/work/packages -v $(pwd)/tesseract52.yaml:/work/tesseract52.yaml cgr.dev/chainguard/melange build --arch=x86_64 tesseract52.yaml
1 change: 1 addition & 0 deletions .github/workflows/push-packages.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@ jobs:
docker run --privileged --rm -v $(pwd)/packages:/work/packages -v $(pwd)/libspatialindex.yaml:/work/libspatialindex.yaml -v $(pwd)/melange.rsa:/work/melange.rsa cgr.dev/chainguard/melange build --signing-key melange.rsa --arch=x86_64 libspatialindex.yaml
docker run --privileged --rm -v $(pwd)/packages:/work/packages -v $(pwd)/mesa22.yaml:/work/mesa22.yaml -v $(pwd)/melange.rsa:/work/melange.rsa cgr.dev/chainguard/melange build --signing-key melange.rsa --arch=x86_64 mesa22.yaml
docker run --privileged --rm -v $(pwd)/packages:/work/packages -v $(pwd)/libjpeg-turbo2.yaml:/work/libjpeg-turbo2.yaml -v $(pwd)/melange.rsa:/work/melange.rsa cgr.dev/chainguard/melange build --signing-key melange.rsa --arch=x86_64 libjpeg-turbo2.yaml
docker run --privileged --rm -v $(pwd)/packages:/work/packages -v $(pwd)/tesseract52.yaml:/work/tesseract52.yaml -v $(pwd)/melange.rsa:/work/melange.rsa cgr.dev/chainguard/melange build --signing-key melange.rsa --arch=x86_64 tesseract52.yaml
- name: Configure AWS credentials
uses: aws-actions/configure-aws-credentials@v3
Expand Down
154 changes: 154 additions & 0 deletions tesseract52.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,154 @@
package:
name: tesseract52
version: 5.2.0
epoch: 2
description: Tesseract Open Source OCR Engine (5.2.0)
copyright:
- license: Apache-2.0

environment:
contents:
keyring:
- https://packages.wolfi.dev/os/wolfi-signing.rsa.pub
repositories:
- https://packages.wolfi.dev/os
packages:
- autoconf
- automake
- build-base
- busybox
- ca-certificates-bundle
- cairo-dev
- cmake
- curl-dev
- expat-dev
- fontconfig-config
- fontconfig-dev
- fribidi-dev
- glib-dev
- harfbuzz-dev
- icu-dev
- leptonica-dev
- libarchive-dev
- libfontconfig1
- libjpeg-turbo-dev
- libxft-dev
- opencl-dev
- pango
- pango-dev
- pkgconf
- pkgconf-dev

data:
- name: langs
items:
afr:
ara:
aze:
bel:
ben:
bul:
cat:
ces:
chi_sim:
chi_tra:
chr:
dan:
deu:
eng:
enm:
epo:
equ:
est:
eus:
fin:
fra:
frk:
frm:
glg:
grc:
heb:
hin:
hrv:
hun:
ind:
isl:
ita:
ita_old:
jpn:
kan:
kat:
khm:
kor:
lav:
lit:
mal:
mkd:
mlt:
msa:
nld:
nor:
osd:
pol:
por:
ron:
rus:
slk:
slv:
spa:
spa_old:
sqi:
srp:
swa:
swe:
tam:
tel:
tgl:
tha:
tur:
ukr:
vie:

vars:
tessdata-version: 4.1.0

pipeline:
- uses: git-checkout
with:
repository: https://github.com/tesseract-ocr/tesseract
tag: ${{package.version}}
expected-commit: 5ad5325a0aa8effc47ca033625b6a51682f82767

# Training data is stored in a separate repository
- uses: git-checkout
with:
repository: https://github.com/tesseract-ocr/tessdata
destination: tessdata-${{vars.tessdata-version}}
tag: ${{vars.tessdata-version}}
expected-commit: 4767ea922bcc460e70b87b1d303ebdfed0897da8

- runs: |
# They have some hardcoded include paths
ln -s /usr/include/pango-1.0 /usr/include/pango
- uses: cmake/configure
with:
opts: -DTESSDATA_PREFIX=/usr/share -DUSE_SYSTEM_ICU=on

- uses: cmake/build

- uses: cmake/install

- uses: strip

subpackages:
- name: tesseract52-dev
pipeline:
- uses: split/dev

- range: langs
name: tesseract52-${{range.key}}
pipeline:
- runs: |
mkdir -p ${{targets.subpkgdir}}/usr/share/tessdata
mv tessdata-${{vars.tessdata-version}}/${{range.key}}.traineddata ${{targets.subpkgdir}}/usr/share/tessdata/

0 comments on commit a8dd35c

Please sign in to comment.